© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Julius Han · Gated SAE for Llama3-8B-Instruct
    5. Llama3-8B-IT
    7. Residual Stream
    9. 25-RES-JH
    11. 9038
    Prev
    Next
    INDEX
    Explanations

    references to bravery and the concept of courage

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    Juliushanhanhan/llama-3-8b-it-res/blocks.25.hook_resid_post
    Features
    65,536
    Data Type
    float32
    Hook Name
    blocks.25.hook_resid_post
    Hook Layer
    25
    Architecture
    gated
    Context Size
    1,024
    Dataset
    Juliushanhanhan/openwebtext-1b-llama3-tokenized-cxt-1024
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    rans
    -0.16
    orrent
    -0.15
    vez
    -0.15
    зÑĭ
    -0.15
    ENCIL
    -0.15
    sWith
    -0.15
    _mob
    -0.14
     uprav
    -0.14
    rze
    -0.14
    oldur
    -0.14
    POSITIVE LOGITS
    ously
    0.28
    ous
    0.26
     facing
    0.19
    æķ¢
    0.19
     enough
    0.18
     face
    0.18
    heart
    0.17
     convictions
    0.17
    -face
    0.17
     courage
    0.17
    Activations Density 0.015%

    No Known Activations